BU5565 Assignment
Carbon prices - Coin" rel="nofollow">integration - Engle Granger procedureSUBMISSION DEADLINE: 07/03/2017, 5pm
Word limit: 1000 words (excludin" rel="nofollow">ing tables).Please in" rel="nofollow">include word count.
1. Download dataset “Assignment.wf1" from MyAberdeen. The workfile
contain" rel="nofollow">ins two time series: the logs of two carbon price series. LEUA is the
log of the price for a pollution permit traded in" rel="nofollow">in the European Union
Emission Tradin" rel="nofollow">ing Scheme; LCER is the log ofa so-called Certified
Emission Reduction permit. The data is at daily frequency and the period
of observation is 01/06/2007-30/12/201 1. This assignment deals with
testin" rel="nofollow">ing if the two series are coin" rel="nofollow">integrated.
2. Describe the Engle Granger coin" rel="nofollow">integration test.
3. Use the full sample first. Perform the Engle Granger coin" rel="nofollow">integration test
implemented in" rel="nofollow">in EViews. Write a short summary of your results and
in" rel="nofollow">include the origin" rel="nofollow">inal EViews output.
4. Perform, for the full sample, the in" rel="nofollow">individual steps of the Engle Granger
procedure (see e.g. the textbook by Walter Enders, page 335, Step 1 and
Step 2, or Lecture 3, Slide 35, Step 1.).1 Write a short summary of your
results and in" rel="nofollow">include origin" rel="nofollow">inal EViews outputs. Compare the results to those
of question 3.
5. Now consider the shorter subsample 01/06/2007-30/06/2010. Redo
question 3 and 4. Use a significance level of 10%. Discuss your results.
1. 1 Hin" rel="nofollow">int: residuals from a regression can be calculated usin" rel="nofollow">ing the Proc menu
in" rel="nofollow">in the Equation object, “Make Residual Series" command. Please note that
you do not need to estimate the error correction model.